Posts: 6,481
Threads: 175
Joined: Jan 2012
Thanks: 39
Given 155 thank(s) in 837 post(s)
I've tried the 1.13 snapshot 18w16a, and I'm rather surprised. First, it's pretty unstable, how could have they botched up so many things that worked well before? Such as shift-click in the furnace UI. And the client very often stumbles, puts back two blocks I've just dug, then removes them again. On the functional side of things, is it just me, or are there waaaayy less ores? I've been mining like crazy and I've got only 2 diamonds so far, normally I'd have at least a full stack by this time. Iron is rather sparse, too, and I can see the coal ore pockets are about half the diameter of the old ones (meaning 1/8 of the coal amount).
Also, I've noticed that Windows builds haven't been working since January, and no-one pinged me. Since I've uninstalled all MSVC versions except for 2017, it wouldn't've worked even if I did run the Jenkins agent. Working on that now...
Posts: 6,481
Threads: 175
Joined: Jan 2012
Thanks: 39
Given 155 thank(s) in 837 post(s)
The builds are working again.
Posts: 108
Threads: 1
Joined: Aug 2016
Thanks: 0
Given 3 thank(s) in 10 post(s)
It's sad to see this project.... I don't wanna say die, but it kinda is dying....
Thanks for that honest comment Yan, wish you good luck!
Posts: 1
Threads: 0
Joined: Jul 2018
Thanks: 0
Given 0 thank(s) in 0 post(s)
Stuff like that happens, Yan. I hope you'll manage to sort things out and push through this slump. I think this project is worth it.
Posts: 949
Threads: 16
Joined: May 2013
Thanks: 21
Given 13 thank(s) in 86 post(s)
Dying? pfft. nothing of that sort. Being busy, possibly with a little burnout though? Maybe.
Also the codebase is huge and unwieldy, so forgetting a lot of how it works is very easy, so it's a bit of a chore getting back into it. :P
Posts: 949
Threads: 16
Joined: May 2013
Thanks: 21
Given 13 thank(s) in 86 post(s)
07-16-2018, 06:51 AM
(This post was last modified: 07-16-2018, 06:51 AM by tigerw.)
Also, the upcoming implementation of the global palette to fully realise blockstates might pose a bit of a problem for 1.13 compatibility. I'm assuming the internals of Cuberite isn't getting rewritten to remove data values (unnecessary abstractions belong to slow languages like Java after all), but the translation in the protocol layer will need to be figured out.